Add a way to query attributes of a single mount instead of having to parse
the complete /proc/$PID/mountinfo, which might be huge.
Lookup the mount by the old (32bit) or new (64bit) mount ID. If a mount
needs to be queried based on path, then statx(2) can be used to first query
the mount ID belonging to the path.
Design is based on a suggestion by Linus:
"So I'd suggest something that is very much like "statfsat()", which gets
a buffer and a length, and returns an extended "struct statfs" *AND*
just a string description at the end."
The interface closely mimics that of statx.
Handle ASCII attributes by appending after the end of the structure (as per
above suggestion). Allow querying multiple string attributes with
individual offset/length for each. String are nul terminated (termination
isn't counted in length).
Mount options are also delimited with nul characters. Unlike proc, special
characters are not quoted.
Link: https://lore.kernel.org/all/CAHk-=wh5YifP7hzKSbwJj94+DZ2czjrZsczy6GBimiogZws=rg@xxxxxxxxxxxxxx/
Signed-off-by: Miklos Szeredi <mszeredi@xxxxxxxxxx>
---
arch/x86/entry/syscalls/syscall_64.tbl | 1 +
fs/internal.h | 5 +
fs/namespace.c | 312 ++++++++++++++++++++++++-
fs/proc_namespace.c | 19 +-
fs/statfs.c | 1 +
include/linux/syscalls.h | 3 +
include/uapi/asm-generic/unistd.h | 5 +-
include/uapi/linux/mount.h | 36 +++
8 files changed, 373 insertions(+), 9 deletions(-)
diff --git a/arch/x86/entry/syscalls/syscall_64.tbl b/arch/x86/entry/syscalls/syscall_64.tbl
index 1d6eee30eceb..6d807c30cd16 100644
--- a/arch/x86/entry/syscalls/syscall_64.tbl
+++ b/arch/x86/entry/syscalls/syscall_64.tbl
@@ -375,6 +375,7 @@
451 common cachestat sys_cachestat
452 common fchmodat2 sys_fchmodat2
453 64 map_shadow_stack sys_map_shadow_stack
+454 common statmnt sys_statmnt
#
# Due to a historical design error, certain syscalls are numbered differently
diff --git a/fs/internal.h b/fs/internal.h
index d64ae03998cc..8f75271428aa 100644
--- a/fs/internal.h
+++ b/fs/internal.h
@@ -83,6 +83,11 @@ int path_mount(const char *dev_name, struct path *path,
const char *type_page, unsigned long flags, void *data_page);
int path_umount(struct path *path, int flags);
+/*
+ * proc_namespace.c
+ */
+int show_path(struct seq_file *m, struct dentry *root);
+
/*
* fs_struct.c
*/
diff --git a/fs/namespace.c b/fs/namespace.c
index de47c5f66e17..088a52043bba 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -69,7 +69,8 @@ static DEFINE_IDA(mnt_id_ida);
static DEFINE_IDA(mnt_group_ida);
/* Don't allow confusion with mount ID allocated wit IDA */
-static atomic64_t mnt_id_ctr = ATOMIC64_INIT(1ULL << 32);
+#define OLD_MNT_ID_MAX UINT_MAX
+static atomic64_t mnt_id_ctr = ATOMIC64_INIT(OLD_MNT_ID_MAX);
static struct hlist_head *mount_hashtable __read_mostly;
static struct hlist_head *mountpoint_hashtable __read_mostly;
@@ -4678,6 +4679,315 @@ SYSCALL_DEFINE5(mount_setattr, int, dfd, const char __user *, path,
return err;
}
+static bool mnt_id_match(struct mount *mnt, u64 id)
+{
+ if (id <= OLD_MNT_ID_MAX)
+ return id == mnt->mnt_id;
+ else
+ return id == mnt->mnt_id_unique;
+}
+
+struct vfsmount *lookup_mnt_in_ns(u64 id, struct mnt_namespace *ns)
+{
+ struct mount *mnt;
+ struct vfsmount *res = NULL;
+
+ lock_ns_list(ns);
+ list_for_each_entry(mnt, &ns->list, mnt_list) {
+ if (!mnt_is_cursor(mnt) && mnt_id_match(mnt, id)) {
+ res = &mnt->mnt;
+ break;
+ }
+ }
+ unlock_ns_list(ns);
+ return res;
+}
+
+struct stmt_state {
+ void __user *const buf;
+ size_t const bufsize;
+ struct vfsmount *const mnt;
+ u64 const mask;
+ struct seq_file seq;
+ struct path root;
+ struct statmnt sm;
+ size_t pos;
+ int err;
+};
+
+typedef int (*stmt_func_t)(struct stmt_state *);
+
+static int stmt_string_seq(struct stmt_state *s, stmt_func_t func)
+{
+ struct seq_file *seq = &s->seq;
+ int ret;
+
+ seq->count = 0;
+ seq->size = min_t(size_t, seq->size, s->bufsize - s->pos);
+ seq->buf = kvmalloc(seq->size, GFP_KERNEL_ACCOUNT);
+ if (!seq->buf)
+ return -ENOMEM;
+
+ ret = func(s);
+ if (ret)
+ return ret;
+
+ if (seq_has_overflowed(seq)) {
+ if (seq->size == s->bufsize - s->pos)
+ return -EOVERFLOW;
+ seq->size *= 2;
+ if (seq->size > MAX_RW_COUNT)
+ return -ENOMEM;
+ kvfree(seq->buf);
+ return 0;
+ }
+
+ /* Done */
+ return 1;
+}
+
+static void stmt_string(struct stmt_state *s, u64 mask, stmt_func_t func,
+ stmt_str_t *str)
+{
+ int ret = s->pos >= s->bufsize ? -EOVERFLOW : 0;
+ struct statmnt *sm = &s->sm;
+ struct seq_file *seq = &s->seq;
+
+ if (s->err || !(s->mask & mask))
+ return;
+
+ seq->size = PAGE_SIZE;
+ while (!ret)
+ ret = stmt_string_seq(s, func);
+
+ if (ret < 0) {
+ s->err = ret;
+ } else {
+ seq->buf[seq->count++] = '\0';
+ if (copy_to_user(s->buf + s->pos, seq->buf, seq->count)) {
+ s->err = -EFAULT;
+ } else {
+ str->off = s->pos;
+ str->len = seq->count - 1;
+ s->pos += seq->count;
+ }
+ }
+ kvfree(seq->buf);
+ sm->mask |= mask;
+}
+
+static void stmt_numeric(struct stmt_state *s, u64 mask, stmt_func_t func)
+{
+ if (s->err || !(s->mask & mask))
+ return;
+
+ s->err = func(s);
+ s->sm.mask |= mask;
+}
+
+static u64 mnt_to_attr_flags(struct vfsmount *mnt)
+{
+ unsigned int mnt_flags = READ_ONCE(mnt->mnt_flags);
+ u64 attr_flags = 0;
+
+ if (mnt_flags & MNT_READONLY)
+ attr_flags |= MOUNT_ATTR_RDONLY;
+ if (mnt_flags & MNT_NOSUID)
+ attr_flags |= MOUNT_ATTR_NOSUID;
+ if (mnt_flags & MNT_NODEV)
+ attr_flags |= MOUNT_ATTR_NODEV;
+ if (mnt_flags & MNT_NOEXEC)
+ attr_flags |= MOUNT_ATTR_NOEXEC;
+ if (mnt_flags & MNT_NODIRATIME)
+ attr_flags |= MOUNT_ATTR_NODIRATIME;
+ if (mnt_flags & MNT_NOSYMFOLLOW)
+ attr_flags |= MOUNT_ATTR_NOSYMFOLLOW;
+
+ if (mnt_flags & MNT_NOATIME)
+ attr_flags |= MOUNT_ATTR_NOATIME;
+ else if (mnt_flags & MNT_RELATIME)
+ attr_flags |= MOUNT_ATTR_RELATIME;
+ else
+ attr_flags |= MOUNT_ATTR_STRICTATIME;
+
+ if (is_idmapped_mnt(mnt))
+ attr_flags |= MOUNT_ATTR_IDMAP;
+
+ return attr_flags;
+}
+
+static u64 mnt_to_propagation_flags(struct mount *m)
+{
+ u64 propagation = 0;
+
+ if (IS_MNT_SHARED(m))
+ propagation |= MS_SHARED;
+ if (IS_MNT_SLAVE(m))
+ propagation |= MS_SLAVE;
+ if (IS_MNT_UNBINDABLE(m))
+ propagation |= MS_UNBINDABLE;
+ if (!propagation)
+ propagation |= MS_PRIVATE;
+
+ return propagation;
+}
+
+static int stmt_sb_basic(struct stmt_state *s)
+{
+ struct super_block *sb = s->mnt->mnt_sb;
+
+ s->sm.sb_dev_major = MAJOR(sb->s_dev);
+ s->sm.sb_dev_minor = MINOR(sb->s_dev);
+ s->sm.sb_magic = sb->s_magic;
+ s->sm.sb_flags = sb->s_flags & (SB_RDONLY|SB_SYNCHRONOUS|SB_DIRSYNC|SB_LAZYTIME);
+
+ return 0;
+}
+
+static int stmt_mnt_basic(struct stmt_state *s)
+{
+ struct mount *m = real_mount(s->mnt);
+
+ s->sm.mnt_id = m->mnt_id_unique;
+ s->sm.mnt_parent_id = m->mnt_parent->mnt_id_unique;
+ s->sm.mnt_id_old = m->mnt_id;
+ s->sm.mnt_parent_id_old = m->mnt_parent->mnt_id;
+ s->sm.mnt_attr = mnt_to_attr_flags(&m->mnt);
+ s->sm.mnt_propagation = mnt_to_propagation_flags(m);
+ s->sm.mnt_peer_group = IS_MNT_SHARED(m) ? m->mnt_group_id : 0;
+ s->sm.mnt_master = IS_MNT_SLAVE(m) ? m->mnt_master->mnt_group_id : 0;
+
+ return 0;
+}
+
+static int stmt_propagate_from(struct stmt_state *s)
+{
+ struct mount *m = real_mount(s->mnt);
+
+ if (!IS_MNT_SLAVE(m))
+ return 0;
+
+ s->sm.propagate_from = get_dominating_id(m, ¤t->fs->root);
+
+ return 0;
+}
+
+static int stmt_mnt_root(struct stmt_state *s)
+{
+ struct seq_file *seq = &s->seq;
+ int err = show_path(seq, s->mnt->mnt_root);
+
+ if (!err && !seq_has_overflowed(seq)) {
+ seq->buf[seq->count] = '\0';
+ seq->count = string_unescape_inplace(seq->buf, UNESCAPE_OCTAL);
+ }
+ return err;
+}
+
+static int stmt_mountpoint(struct stmt_state *s)
+{
+ struct vfsmount *mnt = s->mnt;
+ struct path mnt_path = { .dentry = mnt->mnt_root, .mnt = mnt };
+ int err = seq_path_root(&s->seq, &mnt_path, &s->root, "");
+
+ return err == SEQ_SKIP ? 0 : err;
+}
+
+static int stmt_fs_type(struct stmt_state *s)
+{
+ struct seq_file *seq = &s->seq;
+ struct super_block *sb = s->mnt->mnt_sb;
+
+ seq_puts(seq, sb->s_type->name);
+ if (sb->s_subtype) {
+ seq_putc(seq, '.');
+ seq_puts(seq, sb->s_subtype);
+ }
+ return 0;
+}
+
+static int stmt_sb_opts(struct stmt_state *s)
+{
+ struct seq_file *seq = &s->seq;
+ struct super_block *sb = s->mnt->mnt_sb;
+ char *p, *end, *next, *u = seq->buf;
+ int err;
+
+ if (!sb->s_op->show_options)
+ return 0;
+
+ err = sb->s_op->show_options(seq, s->mnt->mnt_root);
+ if (err || seq_has_overflowed(seq) || !seq->count)
+ return err;
+
+ end = seq->buf + seq->count;
+ *end = '\0';
+ for (p = seq->buf + 1; p < end; p = next + 1) {
+ next = strchrnul(p, ',');
+ *next = '\0';
+ u += string_unescape(p, u, 0, UNESCAPE_OCTAL) + 1;
+ }
+ seq->count = u - 1 - seq->buf;
+ return 0;
+}
+
+static int do_statmnt(struct stmt_state *s)
+{
+ struct statmnt *sm = &s->sm;
+ struct mount *m = real_mount(s->mnt);
+
+ if (!capable(CAP_SYS_ADMIN) &&
+ !is_path_reachable(m, m->mnt.mnt_root, &s->root))
+ return -EPERM;
+
+ stmt_numeric(s, STMT_SB_BASIC, stmt_sb_basic);
+ stmt_numeric(s, STMT_MNT_BASIC, stmt_mnt_basic);
+ stmt_numeric(s, STMT_PROPAGATE_FROM, stmt_propagate_from);
+ stmt_string(s, STMT_MNT_ROOT, stmt_mnt_root, &sm->mnt_root);
+ stmt_string(s, STMT_MOUNTPOINT, stmt_mountpoint, &sm->mountpoint);
+ stmt_string(s, STMT_FS_TYPE, stmt_fs_type, &sm->fs_type);
+ stmt_string(s, STMT_SB_OPTS, stmt_sb_opts, &sm->sb_opts);
+
+ if (s->err)
+ return s->err;
+
+ if (copy_to_user(s->buf, sm, min_t(size_t, s->bufsize, sizeof(*sm))))
+ return -EFAULT;
+
+ return 0;
